Articolul precedent |
Articolul urmator |
233 4 |
Ultima descărcare din IBN: 2024-01-10 12:30 |
SM ISO690:2012 LUPU, Maria. Prospects Overview of the Superconducting Neural Networks. In: Electronics, Communications and Computing, Ed. 12, 20-21 octombrie 2022, Chişinău. Chișinău: Tehnica-UTM, 2023, Editia 12, pp. 98-101. DOI: https://doi.org/10.52326/ic-ecco.2022/EL.07 |
EXPORT metadate: Google Scholar Crossref CERIF DataCite Dublin Core |
Electronics, Communications and Computing Editia 12, 2023 |
|||||||
Conferința "Electronics, Communications and Computing" 12, Chişinău, Moldova, 20-21 octombrie 2022 | |||||||
|
|||||||
DOI:https://doi.org/10.52326/ic-ecco.2022/EL.07 | |||||||
Pag. 98-101 | |||||||
|
|||||||
Descarcă PDF | |||||||
Rezumat | |||||||
The long-term efforts of many research groups have led to the fact that by now a large number of different "learning rules" and architectures of neural networks, their hardware implementations and methods of using neural networks to solve applied problems have been accumulated. These intellectual inventions exist in the form of a "technopark" of neural networks. Each network from the technopark has its own architecture, training rules and solves a certain set of tasks. Moreover, specialized highspeed devices can be created on its basis. There are several levels of alienation of a neural network from a universal computer: from network learning on a universal device and the use of rich possibilities in manipulating a task book, learning algorithms and modifying architecture, to complete alienation without learning and modification capabilities, only the functioning of a trained network |
|||||||
Cuvinte-cheie superconducting neural networks, dynamic processes, physics-based models, deep neural networks |
|||||||
|